An audio signal in association with a video image such as a movie is divided into multichannel signals and recorded on the assumption that it is reproduced by speakers arranged on both right and left sides of a screen and speakers arranged at right and left back positions or both right and left sides of the listener. According to such a method, the position of a sound source in a video image and the position of an acoustic image which is actually heard coincide and a sound field having a more natural extent is established.
However, if such audio signals are listened to by using headphones, an acoustic image is oriented in the head, the direction of a video image and the orienting position of the acoustic image do not coincide, and the acoustic image is oriented to an extremely unnatural position.
Even in case of listening to music which is not accompanied with a video image or the like, different from the case of reproducing the sound from the speakers, the sound is likewise heard in the head and an unnatural sound field is also reproduced.
Therefore, there is considered a method whereby head portion transfer functions (impulse response) from speakers arranged in front of the listener to the right and left ears of the listener are preliminarily measured or calculated and are convoluted into audio signals by digital filters and the resultant audio signals are supplied to headphones. According to such a method, since an acoustic image is oriented to a position out of the head, a sound field near that in case of reproducing the sound from the speakers can be reproduced.
According to such a method, however, although the acoustic image is oriented to the position out of the head, when the listener changes the direction of the head, since the acoustic image is moved together with the motion of the head, in case of the acoustic image accompanied with a video image, a deviation occurs between the direction of the video image and the direction of the acoustic image and the acoustic image is oriented to an unnatural position.
Therefore, there is further considered a method whereby the motion of the head of the listener is detected and coefficients of digital filters are updated in accordance with the motion of the head and the direction of an acoustic image is fixed for a listening environment. According to such a method, the acoustic image is not oriented into the head and even if the head is moved, the acoustic image is not moved, so that an acoustic image that is substantially equivalent to the acoustic image which is reproduced by the speakers can be obtained.
However, if a process for the movement of the head as mentioned above is performed, according to experiments, although the orientation of the acoustic image becomes sharp and the sense of direction of the sound becomes clear, the sense of surround which surrounds the listener and is a feature of a virtual stereophonic sound field reproducing feeling is lost, so that it is not suitable for the signal to reproduce a virtual stereophonic sound field.
In case of updating the coefficients of the digital filters in accordance with the motion of the head, even if the head is slightly moved, the coefficients of the digital filters have to be immediately updated each time, so that a number of high speed product sum arithmetic operating circuits and memories are necessary. Thus, a circuit scale increases and the system becomes extremely expensive.
The invention intends to solve the problems as mentioned above.